The Maneuvering Target Tracking Research Based On VRPF

Maneuvering target tracking has wide applications in the military and civilian fileds as a basic problem in the scientific community. In the past few decades, many researchers had studied deep in it in domestic and foreign countries, and had achieved abundant results.With the progress in science and technology and rapid developments in the modern aerospace technology, the navigating speed and the mobility of goal are higher and higher,various application systems request complexly increasingly to the maneuvering target tracking. Model establishment and the filter method are the core of the target tracking, many models are the nonlinear and non-Gaussian for the maneuvering target tracking, therefore we must use the nonlinear filter for filter, particle filter can solve the nonlinear and non-Gaussian problem better at present.This paper first has carried on the brief review on development course of the nonlinear filter theory, and then introduced the present survey situation of particle filter algorithm in detail, and carried on the brief introduction on survey situation of the maneuvering target tracking algorithm at present. The second chapter has introduced modes of maneuvering target movement , including CV model, CA model and Singer model, the current statistical model, and analyzed the characteristics of the various models. The third chapter has introduced particle filter algorithm and its improvement algorithm in detail. The fourth chapter has introduced interacting multiple model(IMM) target tracking technology, which has been used in actual, and introduced IMM algorithm based on particle filter(PF) in the non-linear non-Gaussian model. The fifth chapter has proposed a new maneuvering target movement model based on the intrinsic coordinate frame, and with particle filter, carried on the comparison with IMM algorithm.Because turn movement is maneuvering frequently in actual, in the situation of unknown turning speed, tracking accuracy of the IMM algorithm has dropped, in view of this question, the sixth chapter proposed the IMM algorithm based on variable turning rate particle filte (VRPF).